Methods

Data included in this package include output, driving data, and supplemental calculated information for a modeling study.<br/>

Methods

We developed a watershed modeling framework that can simulate an array of ecosystem services, including land-to-lake flows of water, sediment, and nutrients, and surface water quality (Figure 1). The framework includes process-based representation of natural and managed ecosystems (Agro- IBIS), hydrologic routing of water, sediment, and nutrients through the surface hydrologic network (THMB), and prediction of Secchi disk depth (a measure of lake transparency), summertime lake total phosphorus (TP) concentration, and the probability of hypereutrophy in each lake (Yahara Water Quality Model).
for detailed description of modeling approach please see:
Motew, M., Chen, X., Booth, E.G. et al. The Influence of Legacy P on Lake Water Quality in a Midwestern Agricultural Watershed. Ecosystems 20, 1468–1482 (2017). https://doi.org/10.1007/s10021-017-0125-0

Methods

The state of Wisconsin is comprised of about 15,000 inland lakes ranging from 0.5 to 53,394 ha (WDNR 2009). Most lakes occur in the northern and eastern part of the state as a result of glaciation. about 3,620 lakes are greater than 20 ha and together comprise about 93% of the state's inland lake surface area (Wisconsin Department of Natural Resources 2009). Wisconsin lakes constitute a wide range of physical and biological characteristics. Wisconsin inland lakes support valuable recreational fisheries for a variety of species, including Walleye (Sander vitreus), Northern Pike (Esox lucius), Muskellunge (Esox masquinongy), Yellow Perch (Perca flavescens), Largemouth Bass (Micropterus salmoides), Smallmouth Bass (Micropterus dolomieu), Lake Sturgeon (Acipenser fulvescens), and a variety of sunfish species (Lepomis spp.).

Methods

This study used the R packages “GLMr” (https://github.com/GLEON/GLMr) and “glmtools” (https://github.com/USGS-R/glmtools) to interact with GLM-AED from the R statistical environment.

Hourly meteorological data were obtained for 2010-2016 from the North American Land Data Assimilation System (NLDAS) and included air temperature, relative humidity, short wave radiation, long wave radiation, wind speed, rain accumulation, and snow accumulation.

Average daily stream discharge and nitrogen and phosphorus loads were obtained from the United States Geological Survey (USGS) National Water Information System (NWIS) water quality database. All streams included in this study are monitored by the USGS. This study used 2014 nutrient loading data over the simulation time period since 2016 nutrient data were not yet available. Only the Yahara River and Pheasant Branch Creek have data on daily nutrient loads; thus, Pheasant Branch nutrient concentrations were also applied to both Dorn Creek and Sixmile Creek, which contribute near equivalent volume to Lake Mendota. Since GLM-AED requires nutrient concentration rather than load, water column concentrations were back calculated using load and discharge.

Daily OC loads (POC and DOC, individually) were estimated for each individual inflow (n=4) from observed weekly measurements of OC concentration (doi: 10.6073/pasta/86094298f5a9085869518372934bf9d7)
using stream-specific linear models (Lathrop et al. 1998). The linear relationship between known log-scaled OC loads and log-scaled discharge for each individual stream was used to predict the OC load for each day in the model simulation time period. Again, water column OC concentrations were back calculated using the load and discharge.

A manual calibration routine was conducted to optimize the goodness-of-fit of various state variables in GLM-AED, especially those known to influence OC and GHG dynamics. We used visual comparison of predictions and observations, as well as a quantitative approach, to assess goodness-of-fit. We calibrated water balance and water temperature to effectively simulate lake level and stratification. We also calibrated DO, Secchi depth, TN, TP, DOC, and CH4. Finally, we calibrated phytoplankton dynamics according to four phytoplankton functional groups, recognizing that phytoplankton contribute substantially to the OC pool.

Methods

See Soranno, P.A., Bissell, E.G., Cheruvelil, K.S., Christel, S.T., Collins, S.M., Fergus, C.E., Filstrup, C.T., Lapierre, J.-F., Lottig, N.R., Oliver, S.K., Scott, C.E., Smith, N.J., Stopyak, S., Yuan, S., Bremigan, M.T., Downing, J.A., Gries, C., Henry, E.N., Skaff, N.K., Stanley, E.H., Stow, C.A., Tan, P.-N., Wagner, T., and Webster, K.E. 2015. Building a multi-scaled geospatial temporal ecology database from disparate data sources: fostering open science and data reuse. Gigascience 4: 28. doi: 10.1186/s13742-015-0067-4 for details on how the observed values were obtained. See Oliver et al. Prediction of lake depth across a 17-state region in the U.S. Submitted to Inland Waters for details on the mixed effects model and how predicted values were generated.

Methods

The spatially explicit Lakes, Uplands, Wetlands Integrator (LUWI) model of the NHLD was used to explore the interactions among climate, watershed connections, hydrology and carbon cycling. See Cardille et al. 2007 and Cardille et al. 2009 for details on the LUWI model. See Van de Bogert (2011) for a discussion of how these model outputs are used in conjunction with LUWI to predict the effects on lake carbon cycling under the current and &quot;no freeze&quot; climate scenarios.The climate data used in this modeling effort, precipitation and temperature, were obtained from Minoqua, Wisconsin, USA from 1948-2000. In order to test the effect of a climate without freezing temperatures on lake water and carbon cycling the current climate was modeled in addition to a &ldquo;no freeze&rdquo; scenario where a minimum air temperature of 0 degrees Celsius was imposed on the model. Note that Van de Bogert (2011) only focuses on the current and &ldquo;no freeze&rdquo; climate scenarios, but these data are representative of four climate scenarios: the current climate (base_minoqua_precip), the scenario where the current climate is artificially floored to zero degrees Celsius (no_below_zero), and the scenarios where the current climate is increased and decreased by 10 degrees Celsius (minus_10_degrees and plus_10_degrees).Furthermore, the temperature and precipitation data that was used for the current climate model runs was broken up into aggregates.The aggregates are the length of the 1948-2000 Minoqua temperature and precipitation data that was used in model runs. A total of seven different aggregates were used for model runs under each of the four climate scenarios. The aggregates include temperature and precipitation data from Minoqua, WI, USA for 1. the complete record from 1948-2000 (1948_2000) 2. the driest year which was 1976 (1976_driest) 3. The wettest year which was 1953 (1953_wettest) 4. the five driest years on record from 1948-2000 (5_driest) 5. the five wettest years on record from 1948-2000 (5_wettest) 6. the five coldest years on record for December, January, and February from 1948-2000 (5_coldest_djf) 7. the five warmest years on record for December, January, and February from 1948-2000 (5_warmest_djf).The volume and timing of precipitation to the region were unchanged between scenarios.Evaporation rates were derived from values obtained from the NTL-LTER study site, Sparkling Lake (46.01, -89.70). Refer to Van de Bogert (2011) for a more complete discussion of model inputs and a discussion of the results of the model output. Documentation: Van de Bogert, M.C., 2011. Aquatic ecosystem carbon cycling: From individual lakes to the landscape. ProQuest Dissertations and Theses. The University of Wisconsin - Madison, United States -- Wisconsin, p. 156.Cardille, J.A., Carpenter, S.R., Coe, M.T., Foley, J.A., Hanson, P.C., Turner, M.G., Vano, J.A., 2007. Carbon and water cycling in lake-rich landscapes: Landscape connections, lake hydrology, and biogeochemistry. Journal of Geophysical Research-Biogeosciences 112.Cardille, J.A., Carpenter, S.R., Foley, J.A., Hanson, P.C., Turner, M.G., Vano, J.A., 2009. Climate change and lakes: Estimating sensitivities of water and carbon budgets. Journal of Geophysical Research-Biogeosciences 114.

Methods

Data were collected from two lakes, Sparkling Lake (46.008, -89.701) and Peter Lake (46.253, -89.504), both located in the northern highlands Lake District of Wisconsin and the Upper Peninsula of Michigan over a 10 day period on each lake in July and August of 2007. Refer to Van de Bogert et al. 2011 for limnological characteristics of the study lakes.Measurements of dissolved oxygen and temperature were made every 10 minutes using multiple sondes dispersed horizontally throughout the mixed-layer in the two lakes (n=35 sondes for Sparkling Lake and n=27 sondes for Peter Lake). Dissolved oxygen measurements were corrected for drift.Conductivity, temperature compensated specific conductivity, pH, and oxidation reduction potential were also measured by a subset of sensors in each lake. Of the 35 sondes in Sparkling Lake, 31 were from YSI Incorporated: 15 of model 600XLM, 14 of model 6920, and 2 of model 6600). The remaining sondes placed in Sparkling Lake were 4 D-Opto sensors, Zebra-Tech, LTD. In Peter Lake, 14 YSI model 6920 and 13 YSI model 600XLM sondes were used.Sampling locations were stratified randomly so that a variety of water depths were represented, however, a higher density of sensors were placed in the littoral rather than pelagic zone. See Van de Bogert et al. 2012 for the thermal (stratification) profile of Sparkling Lake and Peter Lake during the period of observation, and for details on how locations were classified as littoral or pelagic. In Sparkling Lake, 11 sensors were placed within the shallowest zone, 12 in the off-shore littoral, and 6 in each of the remaining two zones, for a total of 23 littoral and 12 pelagic sensors. Similarly, 15 sensors were placed in the two littoral zones, and 12 sensors in the pelagic zone.Sensors were randomly assigned locations within each of the zones using rasterized bathymetric maps of the lakes and a random number generator in Matlab. Within each lake, one pelagic sensor was placed at the deep hole which is used for routine-long term sampling.Note that in Sparkling Lake this corresponds to the location of the long-term monitoring buoy. After locations were determined, sensors were randomly assigned to each location with the exception of the four D-Opto sensor is Sparkling Lake, which are a part of larger monitoring buoys used in the NTL-LTER program. One of these was located near the deep hole of the lake while the other three were assigned to random locations along the north shore, south shore and pelagic regions of the lake. Documentation: Van de Bogert, M.C., Bade, D.L., Carpenter, S.R., Cole, J.J., Pace, M.L., Hanson, P.C., Langman, O.C., 2012. Spatial heterogeneity strongly affects estimates of ecosystem metabolism in two north temperate lakes. Limnology and Oceanography 57, 1689-1700.

Methods

Ice and snow sampling was conducted weekly from 14 January to 30 March, 2010 on Lake Mendota when the ice was safe to walk on. A Kovacs Mark III core drill, manufactured by Ice Coring and Drilling Service (ICDS), Space Science and Engineering Center (SSEC) UW Madison, was used to collect ice cores. Snow depth was also measured at the locations where ice cores were sampled. All measurements were made in centimeters. Blue ice can be defined as the portion of the ice core that is strictly frozen lake water. White ice can be defined as &ldquo;snow ice,&rdquo; which occurs when water rushes through cracks in the ice and soaks the overlying snow, resulting in a mixture of ice and snow that subsequently freezes. Total ice is blue ice + snow ice. Finally, snow depth was calculated as the average of 10 snow depth samples at each sampling location.